Partially Bayesian active learning cubature for structural reliability analysis with extremely small failure probabilities

Research output: Contribution to journalArticleResearchpeer review

Authors

  • Chao Dang
  • Matthias G.R. Faes
  • Marcos A. Valdebenito
  • Pengfei Wei
  • Michael Beer

Research Organisations

External Research Organisations

  • TU Dortmund University
  • Northwestern Polytechnical University
  • University of Liverpool
  • Tongji University
View graph of relations

Details

Original languageEnglish
Article number116828
Number of pages17
JournalComputer Methods in Applied Mechanics and Engineering
Volume422
Early online date9 Feb 2024
Publication statusPublished - 15 Mar 2024

Abstract

The Bayesian failure probability inference (BFPI) framework provides a well-established Bayesian approach to quantifying our epistemic uncertainty about the failure probability resulting from a limited number of performance function evaluations. However, it is still challenging to perform Bayesian active learning of the failure probability by taking advantage of the BFPI framework. In this work, three Bayesian active learning methods are proposed under the name ‘partially Bayesian active learning cubature’ (PBALC), based on a cleaver use of the BFPI framework for structural reliability analysis, especially when small failure probabilities are involved. Since the posterior variance of the failure probability is computationally expensive to evaluate, the underlying idea is to exploit only the posterior mean of the failure probability to design two critical components for Bayesian active learning, i.e., the stopping criterion and the learning function. On this basis, three sets of stopping criteria and learning functions are proposed, resulting in the three proposed methods PBALC1, PBALC2 and PBALC3. Furthermore, the analytically intractable integrals involved in the stopping criteria are properly addressed from a numerical point of view. Five numerical examples are studied to demonstrate the performance of the three proposed methods. It is found empirically that the proposed methods can assess very small failure probabilities and significantly outperform several existing methods in terms of accuracy and efficiency.

Keywords

    Bayesian active learning, Bayesian failure probability inference, Learning function, Small failure probability, Stopping criterion, Structural reliability analysis

ASJC Scopus subject areas

Cite this

Partially Bayesian active learning cubature for structural reliability analysis with extremely small failure probabilities. / Dang, Chao; Faes, Matthias G.R.; Valdebenito, Marcos A. et al.
In: Computer Methods in Applied Mechanics and Engineering, Vol. 422, 116828, 15.03.2024.

Research output: Contribution to journalArticleResearchpeer review

Download
@article{195cc27df34c4caa9019e08088c78fec,
title = "Partially Bayesian active learning cubature for structural reliability analysis with extremely small failure probabilities",
abstract = "The Bayesian failure probability inference (BFPI) framework provides a well-established Bayesian approach to quantifying our epistemic uncertainty about the failure probability resulting from a limited number of performance function evaluations. However, it is still challenging to perform Bayesian active learning of the failure probability by taking advantage of the BFPI framework. In this work, three Bayesian active learning methods are proposed under the name {\textquoteleft}partially Bayesian active learning cubature{\textquoteright} (PBALC), based on a cleaver use of the BFPI framework for structural reliability analysis, especially when small failure probabilities are involved. Since the posterior variance of the failure probability is computationally expensive to evaluate, the underlying idea is to exploit only the posterior mean of the failure probability to design two critical components for Bayesian active learning, i.e., the stopping criterion and the learning function. On this basis, three sets of stopping criteria and learning functions are proposed, resulting in the three proposed methods PBALC1, PBALC2 and PBALC3. Furthermore, the analytically intractable integrals involved in the stopping criteria are properly addressed from a numerical point of view. Five numerical examples are studied to demonstrate the performance of the three proposed methods. It is found empirically that the proposed methods can assess very small failure probabilities and significantly outperform several existing methods in terms of accuracy and efficiency.",
keywords = "Bayesian active learning, Bayesian failure probability inference, Learning function, Small failure probability, Stopping criterion, Structural reliability analysis",
author = "Chao Dang and Faes, {Matthias G.R.} and Valdebenito, {Marcos A.} and Pengfei Wei and Michael Beer",
note = "Funding Information: Chao Dang is mainly supported by China Scholarship Council (CSC) . Pengfei Wei is grateful to the support from the National Natural Science Foundation of China (grant no. 51905430 and 72171194 ). ",
year = "2024",
month = mar,
day = "15",
doi = "10.1016/j.cma.2024.116828",
language = "English",
volume = "422",
journal = "Computer Methods in Applied Mechanics and Engineering",
issn = "0045-7825",
publisher = "Elsevier",

}

Download

TY - JOUR

T1 - Partially Bayesian active learning cubature for structural reliability analysis with extremely small failure probabilities

AU - Dang, Chao

AU - Faes, Matthias G.R.

AU - Valdebenito, Marcos A.

AU - Wei, Pengfei

AU - Beer, Michael

N1 - Funding Information: Chao Dang is mainly supported by China Scholarship Council (CSC) . Pengfei Wei is grateful to the support from the National Natural Science Foundation of China (grant no. 51905430 and 72171194 ).

PY - 2024/3/15

Y1 - 2024/3/15

N2 - The Bayesian failure probability inference (BFPI) framework provides a well-established Bayesian approach to quantifying our epistemic uncertainty about the failure probability resulting from a limited number of performance function evaluations. However, it is still challenging to perform Bayesian active learning of the failure probability by taking advantage of the BFPI framework. In this work, three Bayesian active learning methods are proposed under the name ‘partially Bayesian active learning cubature’ (PBALC), based on a cleaver use of the BFPI framework for structural reliability analysis, especially when small failure probabilities are involved. Since the posterior variance of the failure probability is computationally expensive to evaluate, the underlying idea is to exploit only the posterior mean of the failure probability to design two critical components for Bayesian active learning, i.e., the stopping criterion and the learning function. On this basis, three sets of stopping criteria and learning functions are proposed, resulting in the three proposed methods PBALC1, PBALC2 and PBALC3. Furthermore, the analytically intractable integrals involved in the stopping criteria are properly addressed from a numerical point of view. Five numerical examples are studied to demonstrate the performance of the three proposed methods. It is found empirically that the proposed methods can assess very small failure probabilities and significantly outperform several existing methods in terms of accuracy and efficiency.

AB - The Bayesian failure probability inference (BFPI) framework provides a well-established Bayesian approach to quantifying our epistemic uncertainty about the failure probability resulting from a limited number of performance function evaluations. However, it is still challenging to perform Bayesian active learning of the failure probability by taking advantage of the BFPI framework. In this work, three Bayesian active learning methods are proposed under the name ‘partially Bayesian active learning cubature’ (PBALC), based on a cleaver use of the BFPI framework for structural reliability analysis, especially when small failure probabilities are involved. Since the posterior variance of the failure probability is computationally expensive to evaluate, the underlying idea is to exploit only the posterior mean of the failure probability to design two critical components for Bayesian active learning, i.e., the stopping criterion and the learning function. On this basis, three sets of stopping criteria and learning functions are proposed, resulting in the three proposed methods PBALC1, PBALC2 and PBALC3. Furthermore, the analytically intractable integrals involved in the stopping criteria are properly addressed from a numerical point of view. Five numerical examples are studied to demonstrate the performance of the three proposed methods. It is found empirically that the proposed methods can assess very small failure probabilities and significantly outperform several existing methods in terms of accuracy and efficiency.

KW - Bayesian active learning

KW - Bayesian failure probability inference

KW - Learning function

KW - Small failure probability

KW - Stopping criterion

KW - Structural reliability analysis

UR - http://www.scopus.com/inward/record.url?scp=85184519658&partnerID=8YFLogxK

U2 - 10.1016/j.cma.2024.116828

DO - 10.1016/j.cma.2024.116828

M3 - Article

AN - SCOPUS:85184519658

VL - 422

JO - Computer Methods in Applied Mechanics and Engineering

JF - Computer Methods in Applied Mechanics and Engineering

SN - 0045-7825

M1 - 116828

ER -

By the same author(s)